- Optimize generting random files.
- Reduce big file of 128MiB to 32MiB (git was never made for large files
anyways, but simply tests that it works).
- Reduce looped git operations from 100 iterations to 10.
- Add extra print statements to know what a slow test is doing, this
also helps to see if a particular piece of code in a slow test is the
culprit or if the test is just very extensive.
- Set `[ui.notification].EVENT_SOURCE_UPDATE_TIME` to 1s to speed up
`TestEventSourceManagerRun`.
- Sneaked in some general test improvements.
When the CI vars.ROLE is forgejo-coding, it is assumed to be the
repository where collaborative coding happens,
i.e. https://codeberg.org/forgejo/forgejo
When the CI vars.ROLE is forgejo-testing, it is assumed that only codebase
testing is to be run and no other tests such as release build
integration, label constraints, backporting etc.
When the CI vars.ROLE is forgejo-coding, it is assumed to be the
repository where collaborative coding happens,
i.e. https://codeberg.org/forgejo/forgejo
When the CI vars.ROLE is forgejo-testing, it is assumed that only codebase
testing is to be run and no other tests such as release build
integration, label constraints, backporting etc.
- When a dependency is renamed, specified via `package="actual-name"` in
Cargo.toml, this should become the name of the depedency when the
package is retrieved from the registery by cargo and the old name should
be available in the `package` field.
- The reference implementation also does this: 490e66a9d6/src/controllers/krate/publish.rs (L702-L705)
- Resolves#5936
- Unit test added.
Notify https://code.forgejo.org/forgejo/forgejo that a new release was
published by setting the trigger label to
https://code.forgejo.org/forgejo/forgejo/issues/5.
It is only ever useful when a stable release is published, the
experimental releases are not mirrored. But it is triggered in all
cases. This will waste a few mirror check daily, when experimental
releases are built. This is an improvement compared to the current
situation where mirrors are checked hourly:
* Instead of being checked 24 times per day it will be down to less
than 5
* The mirror happens immediately after the release is published
instead of waiting for the next run of the cron job.
If a mirror operation is in progress, as evidenced by the presence of
the trigger label on the issure, it means two releases are being
published. Wait up to 1h for the mirror to complete and remove the
trigger label.
Goals:
- speedup
- less flakiness
- best practices and more use
- documentation
config:
- sync ports in Makefile and playwright config
(otherwise, some tests fail locally because they assert the full URL including the (wrong) port)
- even more generous timeouts
- limit workers to one again (because I finally understand how
Playwright works)
- allow nested functions to group them together with the related test
all:
- deprecate waitForLoadState('networkidle')
- it is discouraged as per https://playwright.dev/docs/api/class-page#page-wait-for-load-state
- I could not find a usage that seems to require it actually (see
added documentation in README)
- adding an exception should be made explicitly
- it does not do what you might expect anyway in most cases
- only log in when necessary
webauthn:
- verify that login is possible after disabling key
- otherwise, the cleanup was not necessary after the previous refactor to create a fresh user each
issue-sidebar / WIP toggle:
- split into smaller chunks
- restore original state first
- add missed assertion to fix race condition (not waiting
before state was reached)
- explicitly toggle the state to detect mismatch earlier
issue-sidebar / labels:
- restore original state first
- better waiting for background request
Once a month.
* Without throttling multiple test / linter packages are updated every
day and this exceeds the review capacity of Forgejo contributors at
the moment. Some of them even release more than once a day,
apparently on every commit.
* Add @axe-core/playwright
Under the assumption that all dependencies have at least some test
coverage. It may not be always true but it is generally true. Ideally
there would be an inventory, a checklist of dependencies that miss
test coverage, but that does not exist.
